Absolute vs Conditional Convergence

When we add up infinitely many numbers, the sum might converge to a finite value — but not all convergent series are alike. Some series converge in a very robust sense: even if you replace every term with its absolute value (removing all minus signs), the resulting series still converges. Others are more fragile: they only converge because of a delicate cancellation between positive and negative terms. These two situations are called absolute convergence and conditional convergence, respectively, and distinguishing between them is one of the most important ideas in the study of infinite series. The classic example of a conditionally convergent series is the alternating harmonic series 1 − 1/2 + 1/3 − 1/4 + ⋯, which converges to ln 2. However, if you take the absolute values of every term, you get the ordinary harmonic series 1 + 1/2 + 1/3 + 1/4 + ⋯, which famously diverges to infinity. In contrast, the series 1 − 1/4 + 1/9 − 1/16 + ⋯ is absolutely convergent because even without the alternating signs the series of absolute values converges (in this case to π²/6). Understanding the distinction matters greatly in higher mathematics and in applications such as signal processing and physics. A remarkable theorem due to Bernhard Riemann shows that conditionally convergent series are unstable in a startling way: by simply rearranging the order of terms, you can make the series sum to any real number you choose — or even diverge. Absolutely convergent series enjoy no such instability; their sums are the same regardless of how the terms are reordered.

Key takeaways
  • Definition of Absolute Convergence
  • Definition of Conditional Convergence
  • Absolute Convergence Implies Convergence
  • Testing for Absolute Convergence: The Ratio and Root Tests
  • Riemann's Rearrangement Theorem
  • Alternating Series Test (Leibniz Test)

Definition of Absolute Convergence

A series ∑aₙ is called absolutely convergent if the series of absolute values ∑|aₙ| converges. In other words, replace every negative term with its positive counterpart; if the new series converges, the original series is absolutely convergent. Example: ∑(−1)ⁿ/n² is absolutely convergent because ∑1/n² = π²/6 < ∞.

Definition of Conditional Convergence

A series ∑aₙ is called conditionally convergent if ∑aₙ converges but ∑|aₙ| diverges. The series converges only due to cancellation from sign changes. Example: The alternating harmonic series ∑(−1)ⁿ⁺¹/n = 1 − 1/2 + 1/3 − ⋯ converges to ln 2, but ∑1/n diverges, so it is conditionally convergent.

Absolute Convergence Implies Convergence

An important theorem states: if ∑|aₙ| converges, then ∑aₙ also converges. The proof uses the comparison test — since −|aₙ| ≤ aₙ ≤ |aₙ|, the partial sums of ∑aₙ are squeezed and bounded. The converse is false: convergence of ∑aₙ does NOT imply convergence of ∑|aₙ|, as the alternating harmonic series demonstrates.

Testing for Absolute Convergence: The Ratio and Root Tests

The Ratio Test checks absolute convergence directly: compute L = lim|aₙ₊₁/aₙ|. If L < 1, the series is absolutely convergent; if L > 1, it diverges; if L = 1, the test is inconclusive. Similarly, the Root Test uses L = lim|aₙ|^(1/n) with the same criteria. These tests inherently test ∑|aₙ|, so a positive result gives absolute convergence.

Riemann's Rearrangement Theorem

For a conditionally convergent series, given any target value S (or ±∞), there exists a rearrangement of the terms that sums to S. The idea: take enough positive terms to exceed S, then enough negative terms to drop below S, then repeat. Because ∑aₙ⁺ = ∞ and ∑aₙ⁻ = −∞ (where aₙ⁺ and aₙ⁻ are the positive and negative parts), this oscillating strategy can target any value. For absolutely convergent series, every rearrangement gives the same sum — a property called unconditional convergence.

Alternating Series Test (Leibniz Test)

A handy tool for detecting conditional convergence: if bₙ > 0, bₙ is decreasing, and lim bₙ = 0, then ∑(−1)ⁿ⁺¹ bₙ converges. This tells us convergence but says nothing about absolute convergence. Pair this test with checking whether ∑bₙ converges (using comparison, p-series, etc.) to classify the series as absolutely or conditionally convergent.

The Key Equation

∑n=1∞an is absolutely convergent  ⟺  ∑n=1∞∣an∣<∞\sum_{n=1}^{\infty} a_n \text{ is absolutely convergent} \iff \sum_{n=1}^{\infty} |a_n| < \infty
  • a_n — nth term (dimensionless): The general term of the series, which may be positive, negative, or zero.
  • |a_n| — absolute value of nth term (dimensionless): The non-negative version of aₙ; all minus signs are removed.
  • \sum_{n=1}^{\infty} |a_n| < \infty — absolute convergence condition (dimensionless): States that the sum of the absolute values converges to a finite limit.

Worked Example

Problem. Classify the following series as absolutely convergent, conditionally convergent, or divergent: (a) ∑(−1)ⁿ⁺¹ / n³ (b) ∑(−1)ⁿ⁺¹ / √n

(a) Absolutely convergent (p-series p=3>1); (b) Conditionally convergent (Alt. Series Test passes, but ∑1/n diverges)\text{(a) Absolutely convergent (p-series } p=3>1\text{); (b) Conditionally convergent (Alt. Series Test passes, but } \sum 1/\sqrt{n} \text{ diverges)}

Answer. Series (a) is absolutely convergent; series (b) is conditionally convergent.

Figure
Diagram
Partial Sums: Absolute vs Conditional ConvergencenSln2π²/12Σ(−1)ⁿ⁺¹/n (cond.)Σ(−1)ⁿ⁺¹/n² (abs.)2468101214
Origins · 1 minBernhard Riemann and how this was discovered
BR
Bernhard Riemann
1826–1866 · German

Riemann proved the startling Rearrangement Theorem (1854): any conditionally convergent series can be rearranged to converge to any prescribed sum, or to diverge. This result, published in his habilitation thesis, demonstrated a fundamental difference between absolute and conditional convergence and revealed deep subtleties in the concept of infinite sums. Riemann's work also laid foundations for Riemannian geometry, the Riemann integral, and the famous Riemann Hypothesis about the zeros of the zeta function. Riemann completed his entire habilitation dissertation — including the rearrangement theorem and the foundations of Riemannian geometry — in just a few months, yet he was so nervous before his oral defense in 1854 that he almost could not deliver the famous geometry lecture that changed mathematics forever.

  • Alternating Series Test (Leibniz Test) — The primary tool for proving convergence of series with alternating signs; combined with a divergence check on |aₙ|, it identifies conditional convergence.
  • Ratio and Root Tests — These tests directly check convergence of ∑|aₙ|, making them primary methods for establishing absolute convergence.
  • The Harmonic Series — The divergence of ∑1/n is the key fact that makes the alternating harmonic series only conditionally convergent — the central motivating example of this concept.
  • Power Series and Radius of Convergence — Inside the radius of convergence, power series are absolutely convergent; at the boundary, absolute vs conditional convergence must be checked separately for each endpoint.
  • Comparison and Limit Comparison Tests — Used to determine whether ∑|aₙ| converges, which is the decisive step in classifying a series as absolutely convergent, conditionally convergent, or divergent.